In less than a decade, the state of the art in fiber-optic transport systems has progressed from simple point-to-point chains of optically amplified fiber spans to massive networks with hundreds of optically amplified spans connecting transparent add-drop nodes spread over transcontinental distances. Cost reduction has been the primary driver for this transformation, and the primary enabler has been the emergence of the reconfigurable optical add/drop multiplexer (ROADM) as a network element (NE).
Exploiting the inherent wavelength granularity of wavelength-division multiplexing (WDM), an optical add/drop multiplexer (OADM) allows some WDM channels (also referred to as wavelengths) to be dropped at a node, while the others traverse the same node without electronic regeneration. Previously, it was necessary to terminate line systems at each node served, and then regenerate the wavelength signals destined for other nodes. The ability to optically add/drop a fraction of a system's wavelengths at a node was first achieved using fixed OADMs. These were constructed from optical filters, and by enabling wavelengths to optically bypass nodes and eliminate unnecessary regeneration, they provided significant cost savings. However, because traffic growth is inherently unpredictable, it is advantageous for the add-drop capability to be reconfigurable.
ROADMs provide many advantages beyond the savings achieved by optically bypassing nodes. In the future, multi-degree ROADMs with adequate reconfiguration speeds may enable shared-mesh restoration at the optical layer. Shared mesh restoration significantly reduces the number of wavelength channels that must be installed as redundant protection circuits. ROADMs also provide operational advantages. Because ROADMs can be reconfigured remotely, they enable new wavelength channels to be installed by simply placing transponders at the end points, without needing to visit multiple intermediate sites. In addition to these cost-saving benefits, ROADMs will enable new services. For example, if transponders are preinstalled, then new circuits can be provided on-demand. The rapid network reconfiguration provided by ROADMs could also become an enabler of dynamic network services, such as switched video for IPTV. For all of these reasons, ROADMs will continue to have a significant effect on the design of optical networks.
Generally, a ROADM is defined as a NE that permits the active selection of add and drop wavelengths within a WDM signal, while allowing the remaining wavelengths to be passed through transparently to other network nodes. Thus, the simplest ROADM will have two line ports (East and West) that connect to other nodes, and one local port (add/drop) that connects to local transceivers. In today's networks, optical links are typically bidirectional, so each line port represents a pair of fibers. When using conventional local transceivers that can process only a single wavelength at a time, the number of fibers in the add/drop port sets the maximum number of wavelengths that can be added or dropped at a given node.
A ROADM with only two line ports (East and West) is referred to as a two-degree ROADM. Practical networks also have a need for multi-degree ROADMs that can serve more than two line ports. In addition to providing local add/drop of from each of its line ports, the multi-degree ROADM must be able to interconnect any individual wavelength from one line port to another, in a reconfigurable way. The degree of a multi-degree ROADM is equal to the number of line-side fiber pairs that it supports (it does not include the number of fiber pairs used in the add/drop portion of the ROADM).
A full ROADM provides add/drop (de)multiplexing of any arbitrary combination of wavelengths supported by the system with no maximum, minimum, or grouping constraints. A partial ROADM only has access to a subset of the wavelengths, or the choice of the first wavelength introduces constraints on other wavelengths to be dropped. The drop fraction of a ROADM is the maximum number of wavelengths that can be simultaneously dropped, divided by the total number of wavelengths in the WDM signal. If a given add or drop fiber is capable of handling any wavelength, it is said to be colorless. If a given add or drop fiber can be set to address any of the line ports (e.g., east or west for a 2-degree ROADM), it is said to be “steerable.” A NE can be configured such that no single failure that will cause a loss of add/drop service to any two of its line ports.
Carriers wish to deploy systems in the most cost-effective manner possible. Today, it is far more cost-effective to initially deploy the minimal amount of equipment that can smoothly evolve to meet future needs, rather than to deploy a fully loaded system configuration from the very beginning. Currently and for the foreseeable future, transponders make up the dominant cost of a fully loaded optical communication system. If a full set of transponders were included in the initial deployment, then a substantial cost would be incurred before the network had sufficient traffic to support the expense. Therefore, systems are routinely designed to permit incremental deployment of transponders on an as-needed basis. Similar considerations also apply to multiplexers, although the economic drivers are not as strong. In general, modular growth will be supported whenever the additional cost and complication of upgrading to higher capacity in the future is small compared to the financial impact of a full equipment deployment at startup. By designing this pay-as-you-grow approach into ROADMs, the network itself can grow in a cost-effective manner. Traditional networks grow by adding and interconnecting stand-alone line systems, incurring substantial cost and complexity. By using ROADMs that allow for modular deployment of additional ports, network growth can benefit from both the equipment and operational efficiencies of integrating line systems as they are needed into a seamless network. Because networks are deployed over the course of years, carriers prefer to be able to grow the nodes of the network from terminals or amplifiers into multi-degree ROADMs. This not only allows the expense to be spread out over years, it also enables the network designers to respond to unforeseen traffic growth patterns.
FIG. 1 is a schematic of a prior art multi-degree ROADM system 100 (four network degrees are shown). Each network degree is coupled to a pair of optical amplifiers 102, with an input connected to a 1×N optical fan-in device, i.e., a power splitter (PS) or wavelength selective switch (WSS) 104), and an output connected to a N×1 optical fan-out device, i.e., WSS 106. Multiplexed optical signals on input 1081 from network degree 1 are selectively directed via PS/WSS 104 to WSSs 106 and associated outputs 1102, 1103 and/or 1104 for network degrees 2, 3 and/or 4, respectively. In the same manner, multiplexed optical signals on inputs 1082, 1083 and 1084 (network degrees 2, 3 and 4) may be similarly routed to the other network degrees of the system. A plurality of multiplexer/demultiplexer assemblies 1121, 1122, 1123, and 1124 are connected to the WSSs 106 and PS/WSSs 104 for locally adding/dropping wavelengths to/from each network degree 1, 2, 3 and 4 by WSSs 106 and PC/WSSs 104. In this implementation, the add/drop wavelengths cannot be redirected between the network degrees to support dynamic wavelength applications such as bandwidth-on-demand, mesh restoration and wavelength redistribution.
An existing ROADM system for providing dynamic add/drop wavelengths uses a degree for the add/drop wavelengths such that the mux/demux is shared among all the other degrees on the node. Another known approach employs a fiber switch that is disposed between the transponders and the mux/demux to provide a centralized transponder manager such that any transponder can be switched to any add/drop port on any degree. However, none of the previous solutions have proven to be economically practical, and they all suffer from limited scalability.
In view of the above, there exists a need for a new type of multi-degree ROADM system that is specifically adapted for bandwidth-on-demand, mesh restoration or wavelength redistribution applications.